1. 04 Jul, 2018 3 commits
  2. 20 Apr, 2017 1 commit
  3. 22 Mar, 2017 1 commit
  4. 07 Dec, 2016 1 commit
    • Quentin Khan's avatar
      FChebCell/FUnifCell: simplify code · 3ebd26e3
      Quentin Khan authored
      The multipole and local expansion share the same implementation. Their
      types only differ by a template tag parameter. Consquently, the
      `getMultipole`/`getLocal` methods are changed to `get`.
      
      The method change is propageted to the files:
      
        - FChebKernel,
        - FChebSymKernel,
        - FChebTensorialKernel,
        - FAdaptChebKernel,
        - FUnifKernel,
        - FUnifTensorialKernel,
        - FAdaptUnifKernel,
        - utestChebyshevDirectPeriodic,
        - utestInterpolationMultiRhs,
      3ebd26e3
  5. 14 Nov, 2016 1 commit
    • Quentin Khan's avatar
      Update the adaptive Chebyshev kernel to use the new data organisation · 32972f69
      Quentin Khan authored
       - Change FMM operators signature to take advantage of the new
         layout. The operators only take as parameters the data they may need
         instead of the whole cell. For instance (simplified):
      
         void M2M(CellClass* parent, CellClass** children);
      
         becomes
      
         void M2M(multipole_t*  parent_m,    symbolic_data_t* parent_s,
                  multipole_t** children_ms, symbolic_data_t* children_ss);
      32972f69
  6. 12 May, 2016 1 commit
  7. 30 Apr, 2016 1 commit
    • Quentin Khan's avatar
      Adaptive kernel adaptors: fix P2L when using vectorisation · 008c4f7c
      Quentin Khan authored
      The P2L computes particles contributions using vectorized units. Each
      computation result is then reduced to the local expansion.
      
      When the particle count is not a multiple of the vector size,
      computation is done with extra zero-initialised particles. This
      behaviour is problematic with 1/r matrix kernel family when the
      interpolation point is the origin (0,0,0). The extra particles
      contribution is then Not-A-Number and the reduction propagates this to
      the full local expansion.
      
      To mitigate this issue, the last (nb_particles % vector_size) particles
      are computed one by one.
      008c4f7c
  8. 29 Apr, 2016 3 commits
  9. 27 Apr, 2016 1 commit
  10. 18 Apr, 2016 2 commits
  11. 15 Apr, 2016 1 commit
  12. 14 Apr, 2016 1 commit
  13. 08 Apr, 2016 1 commit